Project Summary: We have conducted the largest prospective comparative study of laser amygdalohippocampotomy (SLAH), a minimally invasive procedure ablating the amygdalo-hippocampal complex (AHC) versus selective and standard open resection (OR) epilepsy surgery (N=425+). We demonstrated that SLAH is superior to OR in 1-year cognitive outcomes in adults (including episodic verbal memory & perceptual (semantic) recognition or name retrieval of objects/faces), while achieving comparable seizure control. We will extend our analysis longitudinally in an unprecedented manner (5-10 year outcome) leveraging this unique cohort of well-studied patients including neuropsychological (NP), structural and functional neuroimaging and seizure outcomes. We have also amassed a large sample of cases involving stereotactic laser ablation [SLA] of non-AHC structures. The heterogeneity of these SLA cases and the focal nature of the laser ablation surgical zones create an amazing opportunity for lesion-behavioral analyses. We will expand our neuroimaging analyses to include rigorous voxel lesion symptom mapping (VLSM) and novel methods of connectivity modeling with diffusion tensor imaging and resting state fMRI. We will focus on the neural circuitry responsible for declarative memory and language, while determining the optimal surgical zones to produce the best long-term cognitive and seizure outcomes. VLSM allows us to determine which temporal lobe regions are critical for specific cognitive functions, while connectivity analyses will reveal the interconnectedness of these regions in the individual patient (allowing us to determine the relative risk of decline at the patient level). Our preliminary findings indicate that: 1) SLAH patients exhibit unexpected improvement in multiple cognitive functions, which is unparalleled in their OR counterparts, and 2) SLA procedures sparing the AHC and broader medial TL regions frequently lead to severe decline in verbal memory/naming, particularly in cases of focal ablations of the language dominant temporal pole or fusiform gyrus. This suggests that verbal declarative memory depends upon a broad network extending beyond the AHC, which is a paradigm shift for surgical planning (i.e., regions thought to be expendable during surgical approach are actually critical to function). This knowledge is fundamentally important from mechanistic and clinical translational points of view, and our data and planned analyses provide a unique and feasible way to draw robust conclusions on these long-standing knowledge gaps. Our data raise pivotal questions about existing declarative memory models and the mechanism of recovery underlying the superior cognitive outcomes in many SLAH/SLA cases (e.g., functional sparing vs. reorganization), which we will address in this renewal by following our patients over time. Finally, while more epilepsy centers are employing SLA due to its improved cognitive outcomes, ultimate clinical adoption rests on seizure control, making it critical to determine SLAH efficacy over a longer time span.